xAI model product
Grok 4 Heavy is the multi-agent version of Grok 4, released alongside the standard model in summer 2025.
Updated Aug 12, 2026. Default version: Grok-4 Heavy
Technical details for the model's default version.
This profile uses the model's current scored version. Arena ratings and prices are shown separately.
Benchmark scores for Grok-4 Heavy.
| HMMT25 | 96.7% | 01 | 25 | 100.0% | C | |
| USAMO25 | 61.9% | 02 | 3 | 50.0% | C | |
| AIME 2025 | 100.0% | 04 | 114 | 97.3% | C | |
| LiveCodeBench | 79.4% | 11 | 73 | 86.1% | C | |
| Humanity's Last Exam | 50.7% | 19 | 93 | 80.4% | C | |
| GPQA | 88.4% | 28 | 234 | 88.4% | C |
Preference and agent-evaluation results for the default version.
The default version does not have a matching Arena result yet.
Provider-specific output speed and catalog latency for Grok-4 Heavy. Runtime does not affect the capability score.
No provider-specific speed or latency record is linked to the default version yet.
Output Speed is generated output tokens received per second. Catalog latency is reported separately from observed provider TTFT.
Official vendor API pricing appears first, followed by individual provider offers.
The default version has no current input or output token prices.
Official prices use only the vendor's configured official Provider and positive standard USD PAYG rates. Third-party offers remain explicitly labeled.
Available versions of this model. The score column identifies the version used in the overall ranking.
| Grok-4 Heavy | N/A | 68.2 | N/A | N/A | N/A | No | Proprietary |
Key information about Grok 4 Heavy and its available data.
Grok 4 Heavy is the multi-agent version of Grok 4, released alongside the standard model in summer 2025. This system spawns multiple Grok 4 agents in parallel that work independently on problems and then collaborate by comparing their solutions, similar to a study group.
The agents share insights and tricks they discover, with the system intelligently combining their work rather than simply using majority voting. Grok 4 Heavy uses approximately 10x more test-time compute than regular Grok 4, enabling it to solve significantly more complex problems.
On the Humanities Last Exam, it achieves over 50% accuracy on text-only problems, and it scored a perfect result on the AIME 2025 mathematics competition. The system represents a major advancement in multi-agent AI collaboration and reasoning capabilities.
Data as of 2026-08-11.
Open a comparison with the three ranked models immediately above and below this model.
Recommendations prioritize the same model type and family, then the closest LLMBoard score.
Common questions about Grok 4 Heavy.
A release date is not available for the default version.
No official standard PAYG price is currently available for Grok 4 Heavy.
Grok 4 Heavy was created by xAI.
A context window is not available for the default version.
No. The default version is not marked as having publicly available weights.
No provider offering is currently linked to the default version.
Nearby ranked alternatives include Seed 2.0 Pro, Claude Opus 4.5, Gemini 3.5 Flash Cyber.